Efforts to prevent and cure type 1 diabetes (TIDM) are directed at understanding the mechanisms, which lead to pancreatic islet cell destruction and how to prevent it. Insulin peptide B9-23 is a critical target antigen in the NOD mouse model of TIDM and in human prediabetic and T1DM patients. Vaccination of subjects with either insulin, insulin B chain or an altered peptide ligand of B9-23 are currently undergoing clinical trials in new onset T1DM subjects. These therapies were developed based on the effectiveness of these therapies in NOD mice which have similar but different MHC genes from human subjects. Optimization of the antigen and the protocol to give it may be critically important in discovering the most effective way to prevent diabetes. We propose to develop a humanized mouse model of TIDM in order to develop second generation molecular inhibitors which would potentially be more effective than our current therapeutic approaches. The CW4 human T cell clone responds to insulin peptide B9-23 in the context of DQS. We have established constructs of the T cell receptor and propose to develop a TCR-DQ8 transgenic mouse. The DQ8 background in the two strains we will use have differential expression of DQ8 in dendritic and B cells. We hope to establish a model of TIDM where reactivity to insulin B9-23 in the context of human DQ8 can be explored more fully than is possible in vitro. Our specific aims will be to test whether transgenic TCR-HLA-DQ8 mice develop normal CD4 T cells that can respond to antigen in the context of differential expression of DQ8 on either B cells or on B cells and dendritic cells. Whether these mice develop insulitis and diabetes spontaneously or need additional pertubations such as antigen administration, B7-1 expression, PolylC. or CFA to produce hyperglycemia. Using the known crystal structure of B9-23-HLA-DQ8, we will examine rationally designed peptides of B9-23 to develop inhibitors of this interaction using HLA-DQ8 binding studies, the CW4 T cell clone, other human and transgenic DQ8 clones and hybrids and DQ8- peptide tetramers. As the mice become available and a diabetes model is established, we will test the inhibitors in vitro to determine the efficacy and optimal route, dose and timing of administration. If we are successful in establishing a lead compound, we will then proceed on to preclinical studies to demonstrate safety prior to establishing a protocol for human trials.